22 references to Cl100kBase
Microsoft.ML.Tokenizers (22)
Model\TiktokenTokenizer.cs (22)
1055( "gpt-4-", ModelEncoding.Cl100kBase ), // e.g., gpt-4-0314, etc., plus gpt-4-32k
1056( "gpt-3.5-", ModelEncoding.Cl100kBase ), // e.g, gpt-3.5-turbo-0301, -0401, etc.
1057( "gpt-35-", ModelEncoding.Cl100kBase ), // Azure deployment name
1062( "ft:gpt-4", ModelEncoding.Cl100kBase ),
1063( "ft:gpt-3.5-turbo", ModelEncoding.Cl100kBase ),
1064( "ft:davinci-002", ModelEncoding.Cl100kBase ),
1065( "ft:babbage-002", ModelEncoding.Cl100kBase ),
1083{ "gpt-4", ModelEncoding.Cl100kBase },
1084{ "gpt-3.5-turbo", ModelEncoding.Cl100kBase },
1085{ "gpt-3.5", ModelEncoding.Cl100kBase },
1086{ "gpt-3.5-turbo-16k", ModelEncoding.Cl100kBase },
1087{ "gpt-35", ModelEncoding.Cl100kBase }, // Azure deployment name
1088{ "gpt-35-turbo", ModelEncoding.Cl100kBase }, // Azure deployment name
1089{ "gpt-35-turbo-16k", ModelEncoding.Cl100kBase }, // Azure deployment name
1092{ "davinci-002", ModelEncoding.Cl100kBase },
1093{ "babbage-002", ModelEncoding.Cl100kBase },
1097{ "text-embedding-ada-002", ModelEncoding.Cl100kBase },
1098{ "text-embedding-3-small", ModelEncoding.Cl100kBase },
1099{ "text-embedding-3-large", ModelEncoding.Cl100kBase },
1144{ Phi4ModelName, ModelEncoding.Cl100kBase },
1201case ModelEncoding.Cl100kBase:
1541modelEncoding = ModelEncoding.Cl100kBase;